[PATCH v10 4/5] core: Add kernel message dumper to call on oopses and panics

[Ingo Molnar]

* Artem Bityutskiy <dedekind1 at gmail.com> wrote:

> On Fri, 2009-10-16 at 12:10 +0200, Ingo Molnar wrote:
> > * Simon Kagstrom <simon.kagstrom at netinsight.net> wrote:
> > 
> > > +int kmsg_dump_register(struct kmsg_dumper *dumper)
> > > +{
> > > +	unsigned long flags;
> > > +
> > > +	/* The dump callback needs to be set */
> > > +	if (!dumper->dump)
> > > +		return -EINVAL;
> > > +
> > > +	spin_lock_irqsave(&dump_list_lock, flags);
> > > +
> > > +	/* Don't allow registering multiple times */
> > > +	if (dumper->registered) {
> > > +		spin_unlock_irqrestore(&dump_list_lock, flags);
> > > +
> > > +		return -EBUSY;
> > > +	}
> > > +
> > > +	dumper->registered = 1;
> > > +	list_add(&dumper->list, &dump_list);
> > > +	spin_unlock_irqrestore(&dump_list_lock, flags);
> > > +
> > > +	return 0;
> > > +}
> > > +EXPORT_SYMBOL_GPL(kmsg_dump_register);
> > 
> > 
> > i dont want to bikeshed paint this but this sequence caught my eyes. We 
> > generally do flatter and clearer locking sequences:
> > 
> > int kmsg_dump_register(struct kmsg_dumper *dumper)
> > {
> > 	unsigned long flags;
> > 	int err = -EBUSY;
> > 
> > 	/* The dump callback needs to be set */
> > 	if (!dumper->dump)
> > 		return -EINVAL;
> > 
> > 	spin_lock_irqsave(&dump_list_lock, flags);
> > 
> > 	/* Don't allow registering multiple times */
> > 	if (!dumper->registered) {
> > 		dumper->registered = 1;
> > 		list_add_tail(&dumper->list, &dump_list);
> > 		err = 0;
> > 	}
> > 
> > 	spin_unlock_irqrestore(&dump_list_lock, flags);
> > 
> > 	return err;
> > }
> > EXPORT_SYMBOL_GPL(kmsg_dump_register);
> 
> And while we are on it, I think these extra lines before and after 
> spinlocks are unneeded and even a bit annoying :-)

To me they increase readability quite a bit as it allows me to 
concentrate on just the inner critical section without the distraction 
of the lock/unlock sequence. YMMV.

	Ingo